Enriching Multimodal Sentiment Analysis Through Textual Emotional Descriptions of Visual-Audio Content

The specific details of the AAAI2025 paper "Enriching Multimodal Sentiment Analysis Through Textual Emotional Descriptions of Visual-Audio Content"

Dataset Preparation

MOSI, MOSEI and CH-SIMS download: MMSA

In order to obtain textual emotional descriptions, the source data needs to be preprocessed with OpenFace and openSMILE, please download it yourself.